Conditional independence testing based on a nearest-neighbor estimator of conditional mutual information.
Jakob RungePublished in: CoRR (2017)
Keyphrases
- conditional independence
- nearest neighbor
- conditional mutual information
- mutual information
- bayesian networks
- random variables
- selection criterion
- probability distribution
- graphical models
- directed acyclic graph
- causal models
- structure learning
- learning bayesian networks
- axiomatic characterization
- bayesian network structure
- high dimensional
- possibility theory
- statistical independence
- least squares
- knn
- chain graphs
- probability theory
- maximum likelihood
- markov property
- high dimensional data
- probabilistic inference
- machine learning
- image registration
- data points
- feature vectors
- training set
- conditional independencies
- similarity measure
- image processing
- feature selection